This is a continuation of a prospective study of the ability of ambulatory blood pressure to predict cardiovascular morbidity in patients with mild hypertension, which was first started in 1978. Predictor variables evaluated at entry to the study include clinic and ambulatory blood pressures (including measures of pressure level and variability in different settings), left ventricular mass index (LVMI, determined by echocardiography), renin-sodium profile, and other cardiovascular risk factors (e.g., cholesterol and smoking). During follow-up, blood pressure, treatment status, BMI, and clinical course are evaluated. Outcome measures are definite cardiovascular morbid events, defined as sudden cardiac death, myocardial infarction, stroke, congestive heart failure, and coronary artery revascularization. The main hypotheses to be tested are that ambulatory blood pressure will give a better prediction of outcome than clinic pressure, and that patients with white coat hypertension (defined as a high clinic pressure and normal ambulatory pressure) will be at low risk relative to patients with sustained hypertension. Initial results in 729 patients initially studied between 1978 and 1985 using Cox survival analysis show that the four most significant predictors of morbid events are daytime blood pressure variability, age, male sex, and serum cholesterol. Patients with white coat hypertension appear to be at a level of risk intermediate between normotensives and sustained hypertensives, but the differences are not yet significant. Expansion of the cohort size to include patients evaluated initially between 1985 and 1990 should provide nearly 2,000 patients altogether, which together with the longer follow-up of the initial cohort should provide a sufficient number of morbid events to identify the predictive significance of the different blood pressure measures, and their interaction with other risk factors.